• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Decoding capacitor numbers to find a replacement?

-reverb-

-reverb-

Buying, selling... mainly buying
Hi,

I have these capacitors (the 2 blue ones) in a speaker crossover and I am not sure how to decode the numbers stamped on the part to find a proper replacement. In the sister-speaker, these caps are completely blown - and I mean disintegrated.

Any hints how to determine the characteristic of this cap?

View media item 5391
 
Register to hide this ad
Given this is a speaker question, it might good to have the moderator move it to the Speaker forum for more responses.

To the question, trace out where they are wired into the circuits. I'll bet they're in series with the tweeter and midrange - its a 3-way speaker system, right? The more details provided the better we can help.

If so, they're not capacitors at all - they're designed to absorb excess energy that could blow the smaller drivers. On the speaker where these things are blown, what's the state of the drivers?

Cheers,

David
 
It's probably a varistor, to protect from overload.

https://en.wikipedia.org/wiki/Varistor

anyway, check with the application in the circuit. A varistor is a variable resistor. It probably increases the resistance to protect some driver, or decrease the resistance to by-pass something or shunt some signal to ground.

You need to draw a schematic to confirm this.
 
Hey folks. I appreciate the responses. I would love to be able to draw the schematic but I am afraid that will prove to be beyond my abilities - it is a quite complex crossover. Here's a picture:

View media item 5256
You can't really see them in this picture but the items in question are located on the top-right of the crossover, on the connect strip.

A EE friend of mine at work says they look like "X-Y" capacitors to him. As I understand it, these are usually used to shunt transient voltage spikes on AC lines.

Since it looks like we can't determine a proper replacement from just the part number, would I be able to disconnect the remaining good parts and somehow test them to determine the characteristics and thus find a replacement? I have a capacitance meter but I am not sure how I would test if it is indeed a varistor.

Thanks again.
 
Back to my earlier question: What is the state of the drivers that are somehow connected to the varistors? Do they work? I've tried the numbers on the 2 devices at Mouser with no clear results. The logo looks like AVX but that's about all I can tell. Circuit details will be needed to figure out the specs.

Cheers,

David
 
Sorry Dave... missed that question. As best I can tell, the drivers fed by this crossover are healthy.

The speakers are DCM Timewindow^2. These speakers are in 2 parts - a transmission-line bass unit containing 2 woofers and a removable, sealed "hat" portion containing the high frequency drivers. The portion I am looking at right now is the "hat". It houses 2 midrange drivers and 2 tweeters. The mids are mounted on each side of V-shaped front baffle. The 2 tweeters are coaxially mounted in front of each of the mids. In this regard, pretty typical DCM Timewindow layout.

On the sister speaker that has the blown cap/varistor, the right tweeter seems inactive. I haven't checked the tweeter VC yet to see if the coil is open or not.

I have recapped both hats but I still must resolve this part issue before calling them done. When I serviced the first hat, the cap/varistors appeared fine. I didn't realize I had a problem with the cap/varistor until I opened the second hat for recapping. That's when I found the exploded part.

Still left to do is the bass units - they have their own fairly complex crossover as well!
 
I agree that these are probably not capacitors, but are MOV varistors. They act as an over-voltage clamp for very short transients. Longer transients will cause them to self destruct. Usually they are used along with a fuse so that the fuse opens before the MOV explodes.
I don't recognize the manufacturer's symbol and the numbers aren't coming up on google, so I can't help there. You can just leave them out, since in the absence of an overload, they do nothing (conduct no current) so they won't affect the sound.
You may have to guess at a replacement based on the expected peak voltage and the approximate physical size.

Terry
 
Testing the existing (good) parts would be difficult. You would have to apply a very short (microsecond) pulse of high enough voltage (100 or 200 volts or more) that was current limited and see what voltage the device clamps at. Even then you would have to guess at the energy rating (joules).
You might get away with applying a high voltage DC through a very large resistor so that the current is limited to a safe value and measuring the clamp voltage like you would for a zener. I've never tried it. Maybe try the technique out with a sacrificial part with known parameters and see if it works. Of course most people don't have a power supply capable of reaching the voltages required.

I would look at where they are located in the circuit and estimate the peak voltage that they would be expected to see at that point. Then pick a varistor that clamps somewhat above that voltage. Then go with a similar physical size to get the right energy rating).
(or just leave them out).

Terry
 
It is possible that they are PTC resettable fuses. They are more common in speaker crossovers (I think). A PTC is basically the opposite of a varistor. They have a fairly low resistance (like a few ohms) until they get hot from excessive power, then they go to a much higher resistance (hundreds or thousands of ohms). Once they cool down, they go back to the low resistance.
I wouldn't expect a PTC to blow apart like you describe since they tend to protect themselves by going to the high resistance when hot. This is the opposite of a varistor, which in effect shorts out the overload, absorbing it all until the overload goes away or the device explodes.
A PTC would be in series with the driver while a varistor would be in parallel (shunt).
It should be easy to tell which it is by measuring the ones that are still good with an ohmmeter. If they read a few ohms they are PTC resettable fuses. If they read basically open, they are varistors.

Terry
 
Interesting info Terry. I think you have nailed it. Thanks!

I was looking over the Owner's Manual and this jumped out at me:

"The Time Window^2 contains automatic non-mechanical thermal protection devices which protect the bass drivers, midrange drivers, and the tweeters separately. They will allow clean power of up to 2000 watts peak to pass..."

I can identify everything else in the crossover except these parts so this MUST be their role. Also there are 2 of them in each hat (1 for midrange protection, 1 for tweeter protection) and 1 in the bass crossover for woofer protection. Makes sense to me.

btw: That's not a typo - it does state 2000 watts peak!

I'm going to jumper around the blown one tonight and see if my tweeter comes back to life. Who knows, I might get lucky.
 
Unless you're having some sort of significant problem with the crossover, just replace those electrolytics and go on with life.
 
Hey Sam. I am having trouble.

All the drivers, sans 1 tweeter are healthy but the overall speakers sound horrible. Veiled, lifeless, anemic, distorted sounding - like there is something bleeding off the wattage and adding noise. They are a bear to get into (like most DCM stuff) so while I am there, I want to make sure I address everything that I reasonable can.

The hats sound much better after the recap but I need to figure out if this blown "device" is what's causing my dead tweet. I just got into the first bass cabinet the other night and I predict that it will sound much better after fresh caps also. Getting a parts order ready at the moment.

I think I am heading in the right direction here. One of the 200 uf caps in the hat read 820 uf :eek:

As Bob Uecker would say "Juuuust a bit outside"
 
OK. When I was prepping to jumper the larger of the two blue thingys it disintegrated in my hands. Once the two leads are not touching in any way, all of the midrange signal goes away. If I disconnect one leg of the smaller blue thingys, all of the tweeter signal goes away.
 
I believe those blue things are called a poly-switch.

I was just reading about them on the Polk forums. There are a lot of people saying that as they get old they degrade the sound as they deteriorate. Another problem is that once they have been tripped they increase in resistance.

There are a lot of people that replace them with an equivalent resistor to restore the balance of that leg of the network. Yet there are others that like the protection that the poly-switch is supposed to provide.

Depending on what camp you are in they should be replaced with a fresh poly-switch or an equivalent resistor. I believe the nominal value of an un-tripped poly-switch is about .5 Ohms but this may vary depending on manufacturer.

I know PE was selling them with a few different values.


Here is technical description of how they work from TE.
http://www.te.com/content/dam/te-co...al/01-circuit-protection-psw-fundamentals.pdf
 
Last edited:
-reverb- said:
OK. When I was prepping to jumper the larger of the two blue thingys it disintegrated in my hands. Once the two leads are not touching in any way, all of the midrange signal goes away. If I disconnect one leg of the smaller blue thingys, all of the tweeter signal goes away.
Click to expand...

That is because they are in series with the drivers. They are supposed to go HI-Z in the circuit when they heat up and protect the drivers from overcurrent.
 
Last edited:
Well, glad we got to the bottom of that. Thanks all who helped out.

So what to do now?? I guess there is still no good way to determine what should be used as a replacement, right?

The Timewindow 3 spec sheet says it has "self resetting protection circuit" also. Thinking there may be commonality between the two, I looked at the TW 3 schematic but I don't see a symbol that looks like it would be a polyswitch. Of course there is no schematic for the TW^2 "in the wild" that I am aware of.

PE has polyswitches from 0.5A to 1.35A, but picking anything would just be a WAG on my part.

I hate to button them up without the polyswitch in place but it looks like that may be the only route possible. Fill the hole in the circuit with a resistor as SoundOfSound suggest and move on.
 
Do you know what year these speakers came out?

There is a rhyme and reason to decoding the part number if we can dig up the correct data sheet. I think these poly-switches where made by Raychem, Tyco or TE connectivity. Knowing the year would help to dig up the correct data sheet and decode the numbers on the parts.

Here is an old Tyco/Raychem one but it does not match up to the part exactly.
https://www.westfloridacomponents.com/mm5/graphics/ds8/Raychem-Polyswitch.pdf

EDIT: Are there any numbers on the opposite side of the parts?
 
Last edited:
No additional numbers or symbols anywhere else on the parts. Only markings are those shown in the picture.

From what I have been able to determine, they were manufactured from 1982-1984. I read one thread that the guy purchased his in 1984 and his serial numbers are 617 and 618. My serials are 555 and 556 so best guess is they were built towards the end of the run - 83/84 probably. It is believed that less than 1000 pairs were produced.

I measured the diameter of my polyswitches. The smaller ones are ~16.2 mm (.64") and the larger ones are ~19.2 mm (.75"). Not sure if this helps but the ratings seem to be related to size.
 
You must log in or register to reply here.
Back
Top Bottom